Silicon isotopes in an Archaean migmatite confirm seawater silicification of TTG sources

Madeleine E. Murphy,Jane E. Macdonald,Sebastian Fischer,Nicholas J. Gardiner,Richard W. White,Paul S. Savage
DOI: https://doi.org/10.1016/j.gca.2024.01.018
IF: 5
2024-01-26
Geochimica et Cosmochimica Acta
Abstract:Unraveling ancient melting processes is key to understanding how the earliest, tonalite-trondhjemite-granodiorite (TTG)-dominated continental crust formed from partial melting of amphibolite. Application of silicon isotope analysis to ancient crust reveals that Archaean TTGs exhibit consistently high Si isotope values (δ 30 Si) compared to modern granitoids, attributed to seawater-derived silica introduced by either (a) partial melting of variably silicified basalts or (b) assimilation of authigenic silica-rich marine lithologies in the melt source. However, both mechanisms can introduce highly variable δ 30 Si, conflicting with the strikingly consistent δ 30 Si compositions of Archaean TTGs. This study investigates an alternative model, whereby the distinct mineralogy and chemistry of TTG melt sources impart a distinct silicon isotope composition to the melt, compared with "modern" granitoids. We measured δ 30 Si in component parts (melanosome and leucosome) of an Archaean (2.7 Ga) mafic migmatite and coeval amphibolites and mafic granulites from the Kapuskasing uplift, Canada, to explore how Si isotopes fractionate during incipient TTG melt formation. Our data reveal leucosome (i.e., melt) exhibits consistently high δ 30 Si values compared to a relatively isotopically lighter melanosome (i.e., residuum). We also derive inter-mineral silicon isotope fractionation factors for mineral separates that agree well with those of ab initio estimates for the same minerals and show that the magnitude of equilibrium fractionation between TTG source rock and melt replicates that in Phanerozoic granitoids. We conclude the effects of magmatic differentiation on δ 30 Si have remained consistent throughout Earth history, meaning that Archaean TTGs must require a source isotopically heavier than unaltered basalt, as reflected by our amphibolites and mafic migmatite components. The consistently heavy δ 30 Si of seawater through Earth history, and the high SiO 2 content of amphibolites relative to coeval leucosome-free granulites in our study area, imply seawater silicification is the source of the observed high δ 30 Si. Thus, the consistently heavy Si isotope compositions measured in Archaean melt products define a unique aspect of ancient crust formation: that of the silicification of TTG source rock, implying the intrinsic involvement of a primeval hydrosphere.
geochemistry & geophysics
What problem does this paper attempt to address?